TLDR Founders 2026-11-17
Enterprise AI tokenomics ๐ธ, code review at Linear ๐จโ๐ป, donโt hire like Google ๐ผ
Introducing the Databricks AI Index: The state of enterprise AI tokenomics (12 minute read)
The Databricks AI Index is a public resource that brings the company's insights into real-world adoption and spending into one place. It makes the company's internal evaluation suite permanently public. The index shows how new models compare and enterprise AI adoption rates. It is updated weekly as Databricks evaluates new proprietary and open-weight model releases.
The Frontier Blinks (2 minute read)
Anthropic's newest AI models, released last week as Fable 5.2, cost 20% less to use than the top models before them. Until then, only the smaller AI models were getting cheaper. The new models also run about 30% faster, so the same chip turns out more tokens for the same hourly cost. The price cut roughly matches those savings, which Anthropic could have kept as margin. Expect more cuts like it as models keep getting cheaper to run.
If You're Refounding Your B2B Company for AI, Make Your Product Cheaper. Probably Much Cheaper (6 minute read)
HubSpot is rebuilding its products around AI agents, and plenty of other business software companies are doing the same. Most of their prices were set for people, who stop paying attention once they sign. Agents check the cost of every call and call less when a product costs too much. SaaStr's own agents cut their calls to one vendor by about 90% after its bill rose 40%. Price agent usage apart from human seats, and watch calls per customer per day for the first sign of trouble.
State of App Monetization 2026: How Apps Climb From $500 to $1M a Month (15 minute read)
One in four apps climbs revenue tiers within a year, with paths from $500 to $1 million achievable through strategic adjustments. Key factors include optimizing paywall conversion rates from 2.9% to 6.4%, reducing checkout abandonment from 73% to 57%, and enhancing trial-to-paid conversions from 20% to 31%. Apps that experiment more with features like multi-page flows, pricing strategies, and personalized onboarding see higher growth, emphasizing the importance of targeted testing and adjustments.
Why Your Startup Shouldn't Hire Like Google (2 minute read)
A startup's hiring advantage is that it can find talented people before their resumes make them obvious to everyone else. AI will change how many people a successful company needs, as headcount is not the same as impact. Build a team whose potential you believe in, give them room to grow, and be honest about whether it is working.
Code review at Linear (12 minute read)
Linear expects engineers to continue reviewing code in the near term, but it has started building tooling to group related changes, explain their intent, and make meaningful edits easier to distinguish. It seems likely the amount of manual code review will decrease over time as models and review agents improve. The next evolution of review will make it easier for humans to review specific code when needed, make more PRs mergeable without human review, and elevate the focus of review from mechanical implementation to a higher altitude.
AI Search Meta Analysis - Q4 2026 (18 minute read)
AI is changing the path from discovery to decision-making. AI answers reduce clicks, but recommendations can introduce buyers to brands and shape their shortlist before a website visit. Brands need to show up while buyers are researching to give them the information they need to choose. Measure the journey, make decisions easier, and test your visibility.
The Gamma Model Index: OpenAI takes the lead for the first time (10 minute read)
Anthropic has dominated the Gamma Model Index for knowledge work for most of the year, but OpenAI has provided some real competition in the last few months. The new GPT-6.1 Terra replaced Claude Haiku 5.5 and Sonnet 5.5 for several key jobs in Gamma in October. Gamma's agent handles millions of sessions each week, and the company A/B tests every new model on real users. This post looks at how it makes model decisions to serve millions of knowledge workers around the world.
WorkOS Atlas: A secure home for the internal apps your team builds with AI (Tool)
WorkOS Atlas hosts the internal tools that teammates build with coding agents like Claude Code or Cursor. Each app gets company sign-in, and an admin connects tools like Slack and Snowflake once so credentials stay out of app code. Admins see every running app and its owner, and they can limit sensitive apps to certain groups. Atlas is in early access and is onboarding teams one at a time.
Become a founding member of TLDR's AI community (2 minute form)
TLDR is launching a curated community for AI builders to discuss major news and launches. The best links and comments will be featured in TLDR AI.
Apply here to join.
Bumbition (11 minute read)
Two of the founders behind TypeSafe AI, recently valued at $7.5 billion, spent years out of work before they started it. One made her own clothes and won hackathons, and the other became a top-ranked Hearthstone player. Bumbition is a name for that path, waiting for work that's worth doing instead of always chasing the next goal. The time off brings in ideas from other fields and people from outside your industry. The cost is real too, since skills can fade and the right opportunity may come late or never.
The Interface Is the Org Chart (5 minute read)
AI is reshaping company structures by enhancing customer interactions, as exemplified by AI voice agents at Banco Inter, which streamline customer service. AI's ability to access information and expertise across departments can eliminate traditional organizational limitations. This shift suggests potential changes in company divisions, highlighting a need for executives to reconsider existing departmental constraints.
What are your agents doing all night? (16 minute read)
A question on X drew more than 300,000 views last week by asking what everyone's overnight agents do. Most of the hundred-plus replies described short scheduled checks that finish in minutes. Long runs work when a test or score decides which changes to keep. Coding is harder to score, and one team's overnight builds usually need rework in the morning. That team saves the setup for features that take several days, since it costs too much to prepare for a bug fix.
Get our free, 5-minute newsletter read by 200,000 startup founders, entrepreneurs, and CEOs
Join 390,000 readers for
one daily email